VMware

This Question is Possibly Answered

1 "correct" answer available (10 pts) 2 "helpful" answers available (6 pts)
3 Replies Last post: Mar 8, 2009 6:43 AM by eveane  

Checking datastore filesystem? posted: Mar 5, 2009 4:13 AM

Click to view ScottChapman's profile Enthusiast 103 posts since
Dec 8, 2008

I realize that my host machine has crashed a few times.

Is there anyway to validate the filesystem for the datastores?

Re: Checking datastore filesystem?

1. Mar 7, 2009 7:28 AM in response to: ScottChapman
Click to view nick.couchman's profile Champion 4,969 posts since
Jan 13, 2006
For VMFS3, I believe this is done internally at mount time.

Re: Checking datastore filesystem?

2. Mar 7, 2009 8:19 AM in response to: ScottChapman
Click to view mittell's profile Champion 3,096 posts since
Apr 25, 2006
There aren't any end-user tools to check, so hopefully Nick is right and it's automatic, in which case problems would be most likely be reported in /var/log/vmkwarning or /var/log/vmkernel

Re: Checking datastore filesystem?

3. Mar 8, 2009 6:43 AM in response to: ScottChapman
Click to view eveane's profile Enthusiast 21 posts since
Dec 5, 2006

Hi ScottChapman,

nick.couchman this is correct there is no end user tools to force a vmfs filesystem check.

however mittell is correct saying there is a filesystem check at mont time. thsi check will probe a number of metadata (LVM and VMFS) and will report error at that time and will eitehr mount or not the vmfs. in some casethere is still some error that might be tehre but will not prevent the vmfs to mount. those error will occur at a lter stage when doing filesystem operation like starting a vm, reading a file updating the journal. so you need to track those

However if you do a refresh from the graphical interface on the datastore view, a rescan or a vmkfstools -V from the CLI, the same check will be performed. in this case if the vmfs is should fail to mount but is already mounted and has active I/O it will not be unmounted but report the error in vmkernel.

so the key point is to check your vmkernel logs.

if it report any corruption you should no longer trust the datastore itself as you don;t know and will never know the exact extend of the corruption you are experiencing. this is specialy the case when the vmfs is no longer mounting.

if you realise taht the vmfs is no longer mounting, and you have other esx still running vm from that datastore, then do not stop those vm and start to back up the vm from insed the guest (network based, like converter through rdp). if you need to restart them then restart them from inside the guest, not from vmware.

also contect vm-ware support and provide 1 GB dump from that vmfs (disk device and not the partition) as well 32 mB dump from each extend part of taht datastore.

as mention the check done by vmware is only a check and will not fix.

however vmware can check and may be able to fix the metadata, but this will only help for that part, we cannot check the rest of the disk. so the data recovered should be not trusted (consider the respective vmdk as if they were corruption on a physical disk (use fschk and chkdsk and any application tools whcich can review data integrity.)

hope it help

eveane

VMware Developer

SDKs, APIs, Videos, Learn and much more in the Developer community.

Learn More

Developer Sample Code

Increase your developer productivity with VMware API sample code.

Learn More

VMworld Sessions & Labs

Online access to the latest VMworld Sessions & Labs and online services.

Learn more

Purchase PSO Credits Online

Purchase credits to redeem training and consulting services online.

Buy Now

Community Hardware Software

View reported configurations or report your own.

Learn More

VMware vSphere

Come witness the next giant leap in virtualization.

Register Today

Communities